Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method of compressing a digital signal, comprising: reducing redundancies in the digital signal; scaling a block of samples output from the reducing step by a scaling factor, the scaling including collecting a number of samples to form the block of samples and adaptively determining the number of samples to form the block; quantizing the scaled samples to produce quantized samples, the quantized samples having a resolution of a first number of bits; and adaptively determining the resolution used to quantize the scaled samples by, dequantizing the quantized samples to form dequantized samples, descaling the dequantized samples to form a reconstructed block of samples, storing the reconstructed block of samples in a memory, determining a variation between the reconstructed block of samples and the block of samples, and adjusting one or more of the resolution and the number of samples that form the block based on the variation.
A method for compressing a digital signal includes these steps: First, reduce redundancies in the digital signal. Next, scale a block of samples resulting from the redundancy reduction using a scaling factor. This scaling involves collecting a number of samples to form the block, and the number of samples is adaptively determined. Then, quantize the scaled samples, creating quantized samples with a specific bit resolution. Finally, adaptively determine the resolution used for quantization by: dequantizing the quantized samples, descaling the dequantized samples to form a reconstructed block, storing this block, calculating the variation between the original block and the reconstructed block, and adjusting either the bit resolution or the number of samples per block based on this variation.
2. The method of claim 1 , wherein the reducing step comprises: filtering the digital signal; and decimating output from the filtering step.
The method of compressing a digital signal, including reducing redundancies, scaling blocks of samples, and quantizing, where the step of reducing redundancies involves filtering the digital signal, then decimating the output of the filtering step. This filtering and decimation process is done before scaling and quantization.
3. The method of claim 1 , wherein the number of samples to form the block is fixed.
The method of compressing a digital signal, including reducing redundancies, scaling blocks of samples, and quantizing, where the number of samples used to create each block is a fixed, pre-determined value. Therefore, block size does not change during operation.
4. The method of claim 1 , wherein the resolution is fixed.
The method of compressing a digital signal, including reducing redundancies, scaling blocks of samples, and quantizing, where the bit resolution used for quantization is a fixed, pre-determined value. Therefore, the resolution does not change during operation.
5. The method of claim 1 , further comprising: determining a compression error; compressing the compression error; and sending the compressed compression error over a separate link from a link carrying the quantized samples.
The method of compressing a digital signal, including reducing redundancies, scaling blocks of samples, and quantizing, also includes these steps: Determine a compression error. Compress this error. Then, transmit the compressed error using a separate communication link from the link used to transmit the quantized samples. This provides a method of transmitting error correction information.
6. The method of claim 1 , wherein the digital signal is a complex baseband representation (I/Q) of a radio signal.
The method of compressing a digital signal, including reducing redundancies, scaling blocks of samples, and quantizing, where the digital signal that is being compressed is a complex baseband representation (I/Q) of a radio signal. This means that the signal represents both in-phase and quadrature components of the radio signal.
7. A method of decompressing a compressed digital signal at a decompressor, the method comprising: adaptively determining a resolution of received samples, the received samples having a resolution of a number of bits; dequantizing the received samples; descaling a block of the dequantized samples, the descaling including collecting a number of the dequantized samples to form the block of dequantized samples and adaptively determining the number of samples to form the block; and inserting redundancies into the descaled samples to produce a digital signal having a desired spectrum, wherein the decompressor adaptively determines the resolution by receiving a decompression parameter indicating the resolution from a compressor, the compressor configured to determine the resolution by, determining a variation between a block reconstructed samples formed by reconstructing samples previously sent to the decompressor and a latest sample to send to the decompressor, and adjusting one or more of the resolution and the number of samples that form the block based on the variation.
A method for decompressing a compressed digital signal at a decompressor involves: First, adaptively determine the bit resolution of the received samples. Next, dequantize these received samples. Then, descale a block of these dequantized samples. This descaling includes collecting a certain number of the dequantized samples to form a block, and the number of samples per block is adaptively determined. Finally, insert redundancies into the descaled samples to produce a digital signal with a desired spectrum. The decompressor adaptively determines the resolution by receiving a decompression parameter that indicates the resolution from the compressor. The compressor determines the resolution by calculating a variation between a reconstructed block of samples (formed from previously sent samples) and the latest sample to send, then adjusting the resolution and/or block size based on this variation.
8. The method of claim 7 , wherein the dequantizing dequantizes the received samples based on the decompression parameter indicated in the received samples.
The method of decompressing a compressed digital signal that includes adaptively determining resolution, dequantizing, descaling, and inserting redundancies, where the dequantizing step uses the decompression parameter (contained in the received samples) to dequantize the received samples. The decompression parameter indicates how the samples were originally quantized, enabling correct reconstruction.
9. The method of claim 7 , wherein the descaling descales the block of the dequantized samples by a factor indicated in the received samples.
The method of decompressing a compressed digital signal that includes adaptively determining resolution, dequantizing, descaling, and inserting redundancies, where the descaling step uses a scaling factor that is also indicated in the received samples. This scaling factor allows the decompressor to correctly reverse the scaling applied during compression.
10. The method of claim 7 , wherein the inserting step comprises: performing an inverse decimation process on the descaled samples; and inverse filtering output of the performing step.
The method of decompressing a compressed digital signal that includes adaptively determining resolution, dequantizing, descaling, and inserting redundancies, where the step of inserting redundancies involves performing an inverse decimation process on the descaled samples and then performing an inverse filtering operation on the output of the inverse decimation. This reconstructs the original signal's spectrum.
11. A network element, comprising: circuitry generating a first digital signal; a compressor configured to, reduce redundancies in the first digital signal to produce intermediate compressed samples, scale a block of the intermediate compressed samples by a scaling factor, the scaling including collecting a number of samples to form the block of samples and adaptively determining the number of samples to form the block, quantize the scaled samples to produce quantized samples, the quantized samples having a resolution of a first number of bits, adaptively determine the resolution used to quantize the scaled samples by, dequantizing the quantized samples to form tested dequantized samples, descaling the tested dequantized samples to form a reconstructed block of intermediate compressed samples, storing the reconstructed block of intermediate compressed samples in a memory, determining a variation between the reconstructed block of intermediate compressed samples and the block of the intermediate compressed samples, and adjusting one or more of the resolution and the number of samples that form the block based on the variation, and send the quantized samples over a transport link as a first compressed signal; and a decompressor configured to, receive a second compressed signal over the transport link, adaptively determine a resolution of samples in the second compressed signal, the samples having a resolution of a third number of bits, dequantize the samples in the second compressed signal to increase the third number of bits to n bits, where n is a natural number, descale a block of the dequantized samples, and inserting redundancies into the descaled samples to produce a second digital signal having a desired spectrum.
A network element includes circuitry to generate a first digital signal, a compressor, and a decompressor. The compressor reduces redundancies in the digital signal, scales a block of samples using a scaling factor where the number of samples is adaptively determined. It then quantizes the scaled samples with a specific resolution, which is adaptively determined by dequantizing, descaling to form a reconstructed block, storing the block, calculating the variation between the original block and the reconstructed block, and adjusting the resolution or block size. Finally, the compressor sends the quantized samples over a transport link. The decompressor receives the signal, adaptively determines the resolution of samples, dequantizes to increase the bit resolution, descales a block of the dequantized samples, and inserts redundancies to create a second digital signal.
Unknown
December 30, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.